{"id":"10af4645-e2ff-4ffe-b720-68b9cc27ad62","arxiv_id":"2506.10734","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Deep ALMA observations of 10 Lupus disks detect rare CO isotopologues and N2H+, revealing correlations between gas and dust fluxes and giving gas-to-dust mass ratios of 10 to 100.","lead":"New deep ALMA observations of ten planet-forming disks in the Lupus star-forming region detect rare gas tracers and measure disk gas and dust masses. The results show gas line fluxes track dust emission and yield gas-to-dust mass ratios between 10 and 100.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'similar CO abundances' claim rests on excluding Lupus 3 and on a C18O–N2H+ correlation whose only secure anchors are Lupus 2 and 10; with just two N2H+ detections among the nine used, ionization or sample-size effects are not excluded.","rationale":"The reader identified the ionization confounder as the weakest assumption; my stress test agrees but adds a more concrete statistical fragility: even ignoring ionization, the claimed C18O–N2H+ correlation after excluding Lupus 3 is anchored by only two secure N2H+ detections among the nine sources. The paper is careful and transparent, with public data products, detailed uncertainty treatment, and clearly disclosed outlier exclusion, which is genuine credit. However, the central claim of 'similar CO abundances across this sample' is stronger than the data support, because a single excluded point and two bright disks can drive the p<1% result. This does not invalidate the paper's more robust findings, such as the CO isotopologue correlations with continuum and the gas-to-dust size ratios, but it does mean the headline chemistry conclusion should remain conditional. The reader's CONDITIONAL verdict is appropriate; my concern reinforces it rather than changing it. The proposed leave-two-out test would directly show whether the correlation survives without the two dominant points, and the X-ray partial-correlation check would separate the ionization alternative from the CO-abundance interpretation.","tokens_in":46716,"tokens_out":4632,"duration_ms":58705,"concrete_test":"Recompute the censored Kendall's tau (pymccorrelation) for FC18O(2-1) versus FN2H+(3-2) on the nine sources used in Section 5.1, but additionally remove Lupus 2 and Lupus 10, the two N2H+ detections above 5σ among those nine. If p rises above 5% or tau drops below about 0.3, the reported correlation is carried entirely by two disks and does not establish sample-wide similar CO abundances. As a check on the ionization confounder, if X-ray luminosities are available for all ten targets, test whether the C18O/N2H+ flux ratio correlates with L_X; a significant correlation would indicate that ionization, not CO abundance, drives the trend.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim in the Abstract and Section 5.1 is that, with the exception of Lupus 3, a strong correlation between FC18O and FN2H+ indicates similar CO abundances across the sample. The empirical basis is thinner than the prose suggests. In Table 3, N2H+ is detected above 5σ in only three of ten disks (Lupus 2, 3, and 10); Lupus 1 is marginal at ~3σ, and the other six are upper limits. The correlation is reported only after excluding Lupus 3, so it is computed from nine sources of which only two (Lupus 2 and 10) have secure N2H+ detections. A censored Kendall test can return p<1% in this configuration even if the two bright sources dominate the trend and the remaining seven upper limits contribute little independent information. The physical interpretation is also degenerate: the paper itself notes that N2H+ can be enhanced by X-ray ionization and that Lupus 3 is X-ray bright. If ionization levels vary across the other nine disks, the positive FC18O–FN2H+ trend could track ionization rather than CO abundance or gas mass. The post-hoc nature of the Lupus 3 exclusion, while physically motivated, is not an independent test of the hypothesis. Thus the headline conclusion is not yet robust to alternative statistical treatments or to the ionization confounder that the authors themselves identify.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents ALMA Band 6 and Band 7 observations of ten protoplanetary disks in the Lupus star-forming region as part of the AGE-PRO Large Program. The authors measure continuum flux densities and CO isotopologue (12CO, 13CO, C18O) J=2-1 line fluxes, plus N2H+ J=3-2 fluxes, and derive dust masses, disk radii, and gas masses. The main claims are: (1) strong correlations between CO isotopologue fluxes and continuum flux densities; (2) a strong C18O-N2H+ flux correlation after excluding one disk (Lupus 3), interpreted as similar CO abundances across the sample; (3) gas-to-dust size ratios consistent with a larger Lupus sample; and (4) gas masses estimated from C18O with the Ruaud et al. (2022) grid are consistent with the Trapman et al. (2025a) DALI masses, with gas-to-dust mass ratios of 10-100. The data reduction, flux measurements, and uncertainty treatments (including bootstrapping) are carefully presented, and the authors are transparent about upper limits and cloud contamination.","tokens_in":47032,"tokens_out":4692,"duration_ms":51631,"significance":"If the results hold, this paper provides a valuable homogeneous dataset of deep CO isotopologue and N2H+ observations for disks in a young (1-3 Myr) region, with clear potential for future evolutionary studies when combined with the AGE-PRO Ophiuchus and Upper Sco samples. The careful validation of flux uncertainties, the public release of data products and scripts, and the explicit handling of cloud contamination are strengths. The correlation analyses and the comparison between two thermochemical model grids address an important question about whether C18O can serve as a gas-mass tracer in comparable disks. However, the headline 'similar CO abundances' conclusion is currently not robust to small-number statistics and a plausible ionization confounder, and the model comparison is not a fully independent cross-check. The paper's significance is therefore conditional on strengthening these points.","major_comments":[{"comment":"The claim of a strong F_C18O(2-1) versus F_N2H+(3-2) correlation rests on very sparse detections. In Table 3, N2H+ is detected above 5 sigma in only Lupus 2, 3, and 10, with Lupus 1 at roughly 3 sigma and the remaining six sources as upper limits; after excluding Lupus 3, the correlation is computed from nine points of which only two have secure N2H+ detections. A censored Kendall test can return p<1% in this configuration even when the two bright detections dominate the trend and the upper limits add little independent information. Please report the correlation statistic with and without each secure detection, perform a sensitivity analysis (e.g., jackknife or bootstrap p-values), and state explicitly how many detections drive the result. This is load-bearing because the Abstract and Section 6 point 4 use this correlation to conclude similar CO abundances.","section":"Section 5.1, Figure 11, Table 3"},{"comment":"The interpretation of the F_C18O(2-1)-F_N2H+(3-2) relation as evidence of similar CO abundances assumes that N2H+ emission is controlled by CO-destruction chemistry rather than by variations in ionization. The authors themselves note that N2H+ can be enhanced by X-ray ionization and that Lupus 3, the excluded disk, is X-ray bright. If the remaining nine disks differ in ionization, the positive trend could track ionization rather than CO abundance or gas mass. Please quantify this with available ionization tracers (e.g., the N2D+ and DCO+ fluxes presented in Appendix C) or with a model grid spanning X-ray/cosmic-ray ionization rates, or explicitly weaken the conclusion accordingly.","section":"Section 5.1, N2H+ discussion"},{"comment":"The consistency between gas masses from the Ruaud et al. (2022) grid and the Trapman et al. (2025a) DALI grid is not an independent validation. Both methods use the same C18O (and, for the DALI fits, N2H+) measurements from this paper, and both grids are products of the AGE-PRO collaboration. Additionally, Eq. (3) linearly scales a sparse model track to the observed C18O luminosity and assigns ad hoc factor-of-three uncertainties. Please reframe Figure 14 as a comparison between two thermochemical model grids, specify which input constraints are shared and which are independent (e.g., 13CO versus C18O versus N2H+), and discuss how the ad hoc uncertainty assignment affects the claimed agreement. The present wording in the Abstract and Section 5.3 overstates the degree of independent confirmation.","section":"Section 5.3, Eq. (3), Figure 14"}],"minor_comments":[{"comment":"The statement that C18O is detected in 4 targets with SNR>3 sigma, in 5 with SNR~2-3 sigma, and in 1 with SNR~1 sigma is inconsistent with Table 3, which yields >3 sigma detections for seven sources (Lupus 1, 2, 4, 6, 7, 8, 10).","section":"Section 6, summary item 2"},{"comment":"The sentence 'we have detections of 13CO (2-1) for Lupus 4, 5, and 9, and C18O (2-1) for Lupus 7 and 8' is garbled and contradicts the preceding text and Table 3; please rephrase and check the intended detections.","section":"Section 4.1, paragraph after Figure 9"},{"comment":"The orange and green arrows labeled 'decreasing CO abundance' and 'decreasing Mgas' should be defined in the caption or text with the direction of the trend, since the axes show log fluxes and the arrows are otherwise ambiguous.","section":"Section 5.1, Figure 11"},{"comment":"The bootstrapping uncertainty validation is applied only to the three CO isotopologue cubes, not to N2H+; please state whether this was due to computational cost and how the N2H+ flux uncertainties were validated.","section":"Appendix E"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of the journal and presents a useful dataset. The main concern is that the headline conclusion about uniform CO abundances is built on a small number of N2H+ detections and a post-hoc exclusion, with an acknowledged ionization confounder; this needs either more statistical robustness or a softened claim. The model comparison is valuable but should be framed as a grid comparison rather than an independent check. I expect the revision can address these points without changing the overall observational content."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a solid, useful survey paper that deserves a serious referee, but the headline claim is weaker than it looks. The new ALMA data are real: deep Band 6 CO isotopologue and Band 7 N2H+ observations for 10 Lupus disks, with 13CO detected everywhere and C18O in 7–10 depending on threshold. Detection rates are dramatically better than earlier surveys, the flux measurements are carefully validated (RMS vs bootstrap, cloud contamination in 12CO handled explicitly), and the data products and scripts are public on the AGE-PRO website. That alone is worth the paper.\n\nThe main scientific claim—that the C18O–N2H+ flux correlation indicates similar CO abundances across the sample—is not yet robust. The correlation appears only after excluding Lupus 3, and with only two secure N2H+ detections among the nine remaining sources, a censored Kendall test can be driven by the two bright points. The paper itself notes that N2H+ can be enhanced by X-ray ionization and that Lupus 3 is X-ray bright, so the ionization confounder is live. The gas mass comparison between the Ruaud et al. and Trapman et al. grids is also not an independent check: both use the same data and share authors. These are disclosed, but they limit how much weight the conclusions can carry.\n\nI do not see a load-bearing flaw. The fluxes, sample selection, and literature comparisons are careful, and the paper is honest about the limits of its mass estimates. The issue is that the abstract states the correlation as a result, while the body reports it as a post-exclusion finding. The authors should either soften the claim or add a more robust statistical treatment, e.g., including all sources with a proper treatment of upper limits and testing the ionization hypothesis.\n\nFor whom? Disk observers and astrochemists building gas mass inventories. The data will be reused widely. It deserves a serious referee, but the referee should push on the correlation's robustness. I would accept it with major revisions.","headline":"Strong new ALMA data and careful reductions, but the 'similar CO abundances' claim rests on a post-hoc exclusion and only two secure N2H+ detections; worth a serious referee, though the abstract should be softened or the statistics tightened.","tokens_in":47693,"tokens_out":3294,"would_cite":true,"duration_ms":35782,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Ten Lupus disks share one CO chemistry, with a single outlier, so deep C18O observations can weigh disk gas on their own.","keywords":["protoplanetary disks","astrochemistry","planet formation","millimeter astronomy","submillimeter astronomy","CO isotopologues","N2H+","Lupus star-forming region"],"falsifier":"Compile X-ray luminosities for the nine correlated disks and test whether the N2H+/C18O flux ratio rises with X-ray brightness; a positive trend would show that ionization is driving the correlation. A second, independent gas mass probe, such as HD line emission or dynamical masses in the brightest disks, could check whether the C18O-based masses are correct where they matter most.","tokens_in":46499,"feed_emoji":"🪐","tokens_out":12910,"duration_ms":136256,"temperature":0.7,"pith_summary":"This paper presents deep ALMA Band 6 and Band 7 observations of ten planet-forming disks around M3-K6 stars in the Lupus star-forming region, a sample that represents the roughly 1-3 Myr-old disk population there. Its central claim is that, with one exception, the disks share similar carbon monoxide chemistry: the C18O J=2-1 and N2H+ J=3-2 integrated fluxes are strongly correlated once the X-ray-bright disk Lupus 3 is excluded, which is the signature of disks that differ in gas mass rather than in CO abundance. Building on that, gas disk masses estimated from C18O line fluxes alone, using the self-consistent thermo-chemical models of Ruaud et al. (2022), agree with masses derived by Trapman et al. (2025a) from combined CO isotopologue and N2H+ fitting. Both methods put gas-to-dust mass ratios between 10 and 100, and the two well-resolved disks show gas radii about twice their dust radii, evidence against strong radial drift. If right, the paper establishes that deep C18O observations can serve as a practical gas mass census for young disks around low-mass stars.","feed_headline":"Nine of ten Lupus disks share one CO chemistry","feed_subtitle":"The C18O-N2H+ flux tie means deep C18O observations alone can estimate gas disk masses.","key_machinery":"The load-bearing object is the line pair C18O J=2-1 and N2H+ J=3-2. In disk chemistry, CO is the main destroyer of N2H+, so a disk with less CO in its warm molecular layer lets N2H+ survive; the two line fluxes therefore move oppositely with CO abundance and together with gas mass. The observed positive correlation (Lupus 3 excluded) is the prediction of the Trapman et al. (2022) models for disks of similar CO abundance and different gas masses, and it licenses the paper's mass estimates: C18O luminosities are scaled onto the Ruaud et al. (2022) grid of self-consistent disk models, which iterate density, temperature, and chemistry and include isotope-selective photodissociation and grain-surface CO-to-CO2 conversion, and the results are checked against the DALI grid of Trapman et al. (2025a), where CO abundance is a free parameter. Fluxes and radii come from a curve-of-growth aperture method applied to the images, and correlations are tested with Kendall's tau including upper limits.","core_discovery":"On the paper's own terms, the discovery is a flux-flux relationship with a chemical interpretation: across nine of the ten Lupus disks, the C18O J=2-1 flux rises with the N2H+ J=3-2 flux (Kendall's tau test with p<1% once Lupus 3 is removed), and because CO destroys N2H+, the positive slope places the sample on the branch where line brightness tracks gas mass while the CO abundance stays roughly constant. Lupus 3 breaks the pattern, showing strong N2H+ and weak C18O relative to its dust continuum, and the paper reads it as a disk with depleted CO, consistent with it being the brightest X-ray source in the sample. The same logic extends to dust: CO isotopologue fluxes correlate with the 234 GHz continuum flux density, implying a roughly uniform gas-to-dust ratio across the sample, again with Lupus 3 as the outlier. Gas masses follow from scaling C18O luminosities onto the Ruaud et al. (2022) grid, and they match the DALI-based masses of Trapman et al. (2025a), with the largest differences at the faint end still inside the quoted uncertainties.","pith_inferences":["A test the paper leaves implicit: compare N2H+ (3-2) fluxes against X-ray luminosities across all nine correlated disks. If N2H+ excess tracks X-ray brightness, ionization is driving part of the correlation and Lupus 3 is the bright end of a continuum rather than a chemistry anomaly.","The same two-line recipe could be applied to the AGE-PRO samples in Ophiuchus and Upper Sco, which use identical setups. If the C18O-N2H+ correlation persists across regions, C18O-only observations become a practical age-sequence census of disk gas, with N2H+ reserved for spotting outliers.","If CO-depleted disks like Lupus 3 exist at a rate of order one in ten, surveys that rely on C18O alone will systematically under-report gas masses for that subpopulation, skewing gas-to-dust ratio statistics toward the interstellar value.","The paper's own bootstrapping check shows that V1094 Sco's line fluxes could carry about ten times larger uncertainties than the reported RMS values; downstream mass-luminosity scalings built on these ten points should give the largest disk at least that much weight."],"forward_implications":["Deep C18O J=2-1 observations alone can estimate gas disk masses for comparable young disks around low-mass stars, within the uncertainties of the two model grids.","Earlier Lupus gas masses from Miotello et al. (2017) were too low; the new estimates push gas-to-dust mass ratios to 10-100, near or below the interstellar value for the most massive disks.","The two well-resolved disks confirm a gas-to-dust size ratio near 2, placing them below the drift-dominated regime and consistent with the wider Lupus population.","Disks like Lupus 3 are identifiable as CO-depleted through their high N2H+/C18O ratio, and they would be mis-measured by any C18O-only gas mass recipe.","The strong correlations between CO isotopologue fluxes and continuum flux density give predictable detectability, which can guide the design of future gas surveys in other star-forming regions."],"supporting_citations":[{"why":"Supplies the self-consistent thermo-chemical model grid onto which the paper's C18O (2-1) luminosities are scaled to derive gas disk masses.","marker":"Ruaud et al. (2022)"},{"why":"Provides the DALI-based gas masses and CO abundances, fit from CO isotopologues plus N2H+, that the C18O-only estimates are compared against.","marker":"Trapman et al. (2025a)"},{"why":"Predicts how C18O and N2H+ fluxes move with CO depletion versus gas mass, the interpretive basis for reading the observed correlation.","marker":"Trapman et al. (2022)"},{"why":"The earlier DALI gas mass estimates from shallower data that the new, higher masses supersede.","marker":"Miotello et al. (2017)"},{"why":"Documents that enhanced ionization boosts N2H+ emission, the main caveat to the CO-abundance interpretation.","marker":"Anderson et al. (2019)"},{"why":"Shows the C18O/N2H+ flux combination breaks the CO-abundance gas-mass degeneracy and the ionization sensitivity of N2H+.","marker":"Anderson et al. (2022)"},{"why":"The earlier Band 7 Lupus survey whose continuum and 13CO detections set the target selection criteria.","marker":"Ansdell et al. (2016)"},{"why":"The shallower Band 6 survey whose fluxes and upper limits AGE-PRO's deeper detections are compared against.","marker":"Ansdell et al. (2018)"},{"why":"Already applied the Ruaud grid to older Lupus CO fluxes and found larger gas masses, supporting the new agreement.","marker":"Pascucci et al. (2023)"}],"fun_headline_variants":["C18O flux alone can estimate disk gas masses","Lupus disks reveal uniform CO chemistry","One X-ray bright disk breaks CO trend","CO-N2H+ flux correlation ties disk masses"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The correlation argument assumes that N2H+ J=3-2 emission in the nine non-outlier disks is set by how much CO destroys N2H+, not by differences in how strongly each disk is ionized; if ionization varies from disk to disk, the flux correlation could trace ionization rather than uniform CO abundance.","fun_headline_variants_meta":{"raw":{"variants":["C18O flux alone can estimate disk gas masses","Lupus disks reveal uniform CO chemistry","One X-ray bright disk breaks CO trend","CO-N2H+ flux correlation ties disk masses"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000651,"raw_usage":{"total_tokens":3138,"prompt_tokens":1250,"completion_tokens":1888,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":866,"completion_tokens_details":{"reasoning_tokens":1829}},"tokens_in":866,"tokens_out":1888,"duration_ms":19260,"temperature":1.0,"reasoning_tokens":1829,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T04:19:34.511067+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compile X-ray luminosities for the nine correlated disks and test whether the N2H+/C18O flux ratio rises with X-ray brightness; a positive trend would show that ionization is driving the correlation. A second, independent gas mass probe, such as HD line emission or dynamical masses in the brightest disks, could check whether the C18O-based masses are correct where they matter most.","supporting_citations":[],"review_version":1}